Document Type
Conference Paper
Department/Unit
Department of Computer Science
Title
Linear dependency modeling for feature fusion
Language
English
Abstract
This paper addresses the independent assumption issue in fusion process. In the last decade, dependency modeling techniques were developed under a specific distribution of classifiers. This paper proposes a new framework to model the dependency between features without any assumption on feature/classifier distribution. In this paper, we prove that feature dependency can be modeled by a linear combination of the posterior probabilities under some mild assumptions. Based on the linear combination property, two methods, namely Linear Classifier Dependency Modeling (LCDM) and Linear Feature Dependency Modeling (LFDM), are derived and developed for dependency modeling in classifier level and feature level, respectively. The optimal models for LCDM and LFDM are learned by maximizing the margin between the genuine and imposter posterior probabilities. Both synthetic data and real datasets are used for experiments. Experimental results show that LFDM outperforms all existing combination methods. © 2011 IEEE.
Publication Date
2011
Source Publication Title
2011 IEEE International Conference on Computer Vision
Start Page
2041
End Page
2048
Conference Location
Barcelona, Spain
Publisher
IEEE
DOI
10.1109/ICCV.2011.6126477
Link to Publisher's Edition
http://dx.doi.org/10.1109/ICCV.2011.6126477
ISBN (print)
9781457711015
APA Citation
Ma, A., & Yuen, P. (2011). Linear dependency modeling for feature fusion. 2011 IEEE International Conference on Computer Vision, 2041-2048. https://doi.org/10.1109/ICCV.2011.6126477